Seismotectonics of the Arthur's Pass Earthquake of 18 June 1994

Abstract

The magnitude 6.7 Arthur’s Pass earthquake of 18 June 1994 was the largest shock to occur in the Southern Alps for 65 years. We have used data recorded by the New Zealand National Seismograph Network, overseas seismograph stations and sixteen portable seismographs installed in the Arthur’s Pass region immediately after the earthquake to determine the type of fault which took place during the mainshock. No surface faulting was observed for this earthquake. The seismological data indicates that the earthquake mainly involved thrusting on a fault oriented northeast-southwest. Such thrusting is broadly consistent with ground movements resulting from the earthquake measured using Global Positioning System (GPS) satellite receivers. This event is thus taking up the convergent component of the relative motion between the Pacific and Australian plates in the central South Island – events such as this contribute to the uplift that has produced the Southern Alps. The thrusting mechanism for the earthquake determined in this study differs significantly from the mechanisms determined by overseas agencies soon after the event. For at least two years following the event it was not known whether it was a thrust or strike-slip event, let alone on which fault it occurred. This fact is worth noting at a time when the determination of the focal mechanisms of moderate sized earthquakes globally is considered fairly routine.

Definition of the fault plane of the Arthur’s Pass earthquake was rendered difficult because of the unusual distribution of aftershocks of the event. These extend 12 km north-northwest and 30km south-southeast of the actual fault plane. We have sought to explain this unusual distribution of aftershocks by calculating the changes in the stress field in the region caused by fault slip during the mainshock and the largest aftershock (a magnitude 6.0 event near the Harper fault). Such stress changes can bring some regions closer to failure (leading to more aftershocks), and move other regions away from failure. To calculate these stress changes we first determine the regional stress field using geodetic results, earthquake mechanisms and a new method which used P-wave polarity data. All results point to a stress field favouring strike-slip faulting, not thrusting. Thus the thrusting which occurred in the Arthur’s Pass earthquake may be relatively rare. When stress changes caused by the mainshock and largest aftershock are added to the regional stress field, we find a fair correspondence between the aftershock distribution and regions where failure stress has increased. In particular, the subsequent magnitude 6.2 Cass earthquake of 24 November 1995, 30 km to the east, appears to have been triggered by the Arthur’s Pass earthquake.